Suppose you have a thousand-page book, but each page has only a single line of text. You’re supposed to extract the information contained in the book using a scanner, only this particular scanner systematically goes through each and every page, …
With more than 9 billion gigabytes of information traveling the internet every day, researchers are constantly looking for new ways to compress data into smaller packages. Cutting-edge techniques focus on lossy approaches, which achieve compression by intentionally “losing” information from …
How do brains learn? It’s a mystery, one that applies both to the spongy organs in our skulls and to their digital counterparts in our machines. Even though artificial neural networks (ANNs) are built from elaborate webs of artificial neurons, …
On June 27, 2010, the FBI arrested 10 Russian spies who lived and worked as American professionals near New York City. The case, which unraveled an intricate system of false identities and clandestine meetings, exposed one of the largest spy …
Nora Kassner suspected her computer wasn’t as smart as people thought. In October 2018, Google released a language model algorithm called BERT, which Kassner, a researcher in the same field, quickly loaded on her laptop. It was Google’s first language …
Computation is a familiar concept most of us understand intuitively. Take the function f(x) = x + 3. When x is three, f(3) = 3 + 3. Six. Easy. It seems obvious that this function is computable. But some functions …
Machine learning models are incredibly powerful tools. They extract deeply hidden patterns in large data sets that our limited human brains can’t parse. These complex algorithms, then, need to be incomprehensible “black boxes,” because a model that we could crack …
As an undergraduate at the University of Chile, Bernardo Subercaseaux took a dim view of using computers to do math. It seemed antithetical to real intellectual discovery. “There’s some instinct or gut reaction against using computers to solve your problems, …
Despite the wild success of ChatGPT and other large language models, the artificial neural networks (ANNs) that underpin these systems might be on the wrong track. For one, ANNs are “super power-hungry,” said Cornelia Fermüller, a computer scientist at the …
Since the very first days of computer science — a field known for its methodical approach to problem-solving — randomness has played an important role. The first program to run on the world’s first general-purpose electronic computer used randomness to …
Bob Metcalfe has always been a believer in the power of networking. In the 1980s and 1990s he helped popularize the idea that a network’s value grows rapidly with the number of users, a precept now known as Metcalfe’s law. …
On Sunday, February 5, Olof Sisask and Thomas Bloom received an email containing a stunning breakthrough on the biggest unsolved problem in their field. Zander Kelley, a graduate student at the University of Illinois, Urbana-Champaign, had sent Sisask and Bloom …
What movie do these emojis describe? That prompt was one of 204 tasks chosen last year to test the ability of various large language models (LLMs) — the computational engines behind AI chatbots such as ChatGPT. The simplest LLMs produced …
Once in a while, a person can take an abstract concept that’s seemingly too vague for formal study and offer an elegant formal definition. Claude Shannon did it with information, and Andrey Kolmogorov did it with randomness. For the past …
Machine learning is having a moment. Yet even while image generators like DALL·E 2 and language models like ChatGPT grab headlines, experts still don’t understand why they work so well. That makes it hard to understand how they might be …
Lek-Heng Lim longs for a renaissance that would reunite pure and applied mathematics. He points out that the distinction, which in modern math appears to be elemental, is in fact recent. “This break between pure and applied math happened in …
The world has learned two things in the past few months about large language models (LLMs) — the computational engines that power programs such as ChatGPT and Dall·E. The first is that these models appear to have the intelligence and …
In the fourth century, the Greek mathematician Pappus of Alexandria praised bees for their “geometrical forethought.” The hexagonal structure of their honeycomb seemed like the optimal way to partition two-dimensional space into cells of equal area and minimal perimeter — …
I have a confession: When I was 4, I was kicked out of ballet class for asking too many questions. I asked why a lot, like any kid. And I guess the teacher didn’t appreciate it. Luckily for me, the …
Artificial intelligence researchers have celebrated a string of successes with neural networks, computer programs that roughly mimic how our brains are organized. But despite rapid progress, neural networks remain relatively inflexible, with little ability to change on the fly or …